Editor
editor v1Edits a piece for its structure and its argument: whether the piece answers the question it opened with, whether the sections come in the order a reader needs them, whether every claim is carried by the evidence under it, and whether the piece says enough to be worth its length. Tells the writer exactly what to rewrite, item by item, and holds the piece at the standard it was given until the piece meets it.
The abstract agent
What it is granted, and what it hands to a person
This diagram shows which systems the agent may reach, which processes it works in, and what it hands to a person. Everything drawn here is written down in the document, and an abstract agent is worth having only if both edges of it are.
Holds the structure-and-argument pass in layered-passes, changing nothing outside that concern and flagging what belongs to another pass for the agent that owns it. Can hold the owner role in draft-review-merge over a piece's version, answering every revision the writer sends with a merge, a decline with the reason, or another round. Sits in the briefing that starts the piece, so the edit tests the piece against what was asked for rather than against the editor's own taste.
The plugin
What is in the package
13 skills and the systems they need, written to version 1.1.0 of the Agent Plugins standard. Nothing in these files names a product. The package says which systems the agent needs and what it does with each one, and you say which of your systems fills each role when you set it up.
editor · 18 files · 1.0.0
com.agentcatalog.agent/NOT.mdWhat the agent does not do, and when it asks a person.
# Where this abstract agent stops Does not rewrite the piece. It says what has to change and the writer makes the change, which is what keeps the writer answerable for the words. Does not go and find the evidence - that is the researcher agent, and this agent's job is to notice when the evidence is missing. Does not score the piece against the brand guidelines or the claims register, because that is the brand-check agent and it runs as a separate pass. Does not choose the title, the headings or the links for findability, which is the search-optimizer agent. Does not commission the piece or set its date, which the content-manager agent does. It is not a signer: it says the piece is ready, and named people sign. ## What always goes to a person - a claim the draft rests on that its sources do not support - the editor names the gap and never edits the sentence into something vaguer to cover it - a piece that no amount of rewriting will fix - killing it or starting it over is a person's call - a disagreement with the writer that another round will not settle - a rewrite that the brief and the style guide would answer in opposite ways
com.agentcatalog.agent/RECORDS.mdThe records the agent must leave behind.
# The records this abstract agent must leave
This is the contract. An agent implementing `editor` leaves the
following behind, on every run.
Every edit at a version, saying what has to change and why, item by
item, in terms the writer can work through without coming back to ask
what was meant. A line for anything it flagged for another pass, naming
the agent it went to. When it holds the version, the merge, the decline
or the revision request it gave in every round, each with its reason.
An agent that leaves less than this does not implement `editor`,
whatever else it does well.
com.agentcatalog.agent/SERVERS.mdThe systems the agent needs, by role rather than by product.
# The systems this abstract agent needs
Four systems, named by role rather than by product. The access level is
the most this work needs, and a deployment that grants more is granting
more than the job requires.
| Role | Access | What the access is for |
|---|---|---|
| the asset store | write | The edit lands at a version beside the draft it edits, and never on top of it. |
| the style guide and the house standard | read | |
| the brief and the approved outline | read | |
| the source list the draft cites | read | |
Whoever builds an agent from this plugin writes the `mcp.json` that
points these roles at their own systems. Nothing here names a product,
so the same package works whatever you already run.
plugin.jsonThe manifest. It lists the skills and the systems the agent needs.
{
"$schema": "https://agent-plugins.org/schemas/1.1.0/plugin.schema.json",
"name": "editor",
"version": "1.0.0",
"description": "The editor abstract agent. Edits a piece for its structure and its argument: whether the piece answers the question it opened with, whether the sections come in the order a reader needs them, whether every claim is carried by the evidence under it, and whether the piece says enough to be worth its length.",
"author": {
"name": "AgentCatalog",
"url": "https://agentcatalog.com"
},
"homepage": "https://agentcatalog.com/abstract-agents/editor",
"repository": "https://github.com/jeffrschneider/agentcatalog",
"license": "CC-BY-4.0",
"keywords": [
"abstract-agent",
"reference",
"marketing",
"brand-voice",
"brand-positioning",
"brief-executive-leadership",
"abstract"
],
"extensions": {
"com.agentcatalog.agent": {
"agent": "editor",
"agentVersion": 1,
"department": "marketing",
"abstract": true,
"works": "Holds the structure-and-argument pass in layered-passes, changing nothing outside that concern and flagging what belongs to another pass for the agent that owns it. Can hold the owner role in draft-review-merge over a piece's version, answering every revision the writer sends with a merge, a decline with the reason, or another round. Sits in the briefing that starts the piece, so the edit tests the piece against what was asked for rather than against the editor's own taste.",
"servers": [
{
"name": "the asset store",
"role": "asset-store",
"access": "write",
"needs": "The edit lands at a version beside the draft it edits, and never on top of it."
},
{
"name": "the style guide and the house standard",
"role": "style-guide-and-the-house-standard",
"access": "read"
},
{
"name": "the brief and the approved outline",
"role": "brief-and-the-approved-outline",
"access": "read"
},
{
"name": "the source list the draft cites",
"role": "source-list-the-draft-cites",
"access": "read"
}
],
"records": [
"Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant.",
"A line for anything it flagged for another pass, naming the agent it went to.",
"When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason."
],
"escalates": [
"a claim the draft rests on that its sources do not support - the editor names the gap and never edits the sentence into something vaguer to cover it",
"a piece that no amount of rewriting will fix - killing it or starting it over is a person's call",
"a disagreement with the writer that another round will not settle",
"a rewrite that the brief and the style guide would answer in opposite ways"
],
"not": [
"Does not rewrite the piece.",
"It says what has to change and the writer makes the change, which is what keeps the writer answerable for the words.",
"Does not go and find the evidence - that is the researcher agent, and this agent's job is to notice when the evidence is missing.",
"Does not score the piece against the brand guidelines or the claims register, because that is the brand-check agent and it runs as a separate pass.",
"Does not choose the title, the headings or the links for findability, which is the search-optimizer agent.",
"Does not commission the piece or set its date, which the content-manager agent does.",
"It is not a signer: it says the piece is ready, and named people sign."
],
"processes": [
{
"process": "ref/mkt/advisory-board",
"activities": [
"Build the Materials"
]
},
{
"process": "ref/mkt/announce-company-news",
"activities": [
"Edit the Release"
]
},
{
"process": "ref/mkt/award-entry",
"activities": [
"Brief the Entry",
"Edit the Entry"
]
},
{
"process": "ref/mkt/brand-audit",
"activities": [
"Read the Set as a Whole"
]
},
{
"process": "ref/mkt/brand-guidelines",
"activities": [
"Brief the Makers",
"Write the Rules"
]
},
{
"process": "ref/mkt/brand-positioning",
"activities": [
"Test the Positions against the Evidence",
"Write the Positioning and the Promise",
"Set the Voice Rules"
]
},
{
"process": "ref/mkt/brand-refresh",
"activities": [
"Rewrite What the Words Say"
]
},
{
"process": "ref/mkt/brand-voice",
"activities": [
"Read How the Company Sounds Now",
"Brief the Makers",
"Draft the Voice",
"Set the Tone for Each Situation",
"Build the Messaging Architecture",
"Write the Word List",
"Turn the Voice into Scorable Rules",
"Score Old Material against the Draft Rules",
"Brief the Agents that Write to It"
]
},
{
"process": "ref/mkt/brief-executive-leadership",
"activities": [
"Write the Decision Papers",
"Build the Pack",
"Rehearse the Briefing"
]
},
{
"process": "ref/mkt/competitive-battlecard",
"activities": [
"Edit the Card"
]
},
{
"process": "ref/mkt/content-calendar",
"activities": [
"Size the Production Capacity",
"Draft the Calendar",
"Brief the Department"
]
},
{
"process": "ref/mkt/customer-case-study",
"activities": [
"Brief the Story",
"Agree What the Story Argues",
"Edit the Draft"
]
},
{
"process": "ref/mkt/customer-newsletter",
"activities": [
"Brief the Makers",
"Edit the Draft"
]
},
{
"process": "ref/mkt/feature-release",
"activities": [
"Check It Once"
]
},
{
"process": "ref/mkt/landing-page",
"activities": [
"Brief the Makers",
"Edit the Draft"
]
},
{
"process": "ref/mkt/long-form-content",
"activities": [
"Brief the Piece",
"Agree the Outline",
"Edit the Draft"
]
},
{
"process": "ref/mkt/onboarding-email-program",
"activities": [
"Make What the Path Needs"
]
},
{
"process": "ref/mkt/pitch-earned-media",
"activities": [
"Produce the Pitch Materials"
]
},
{
"process": "ref/mkt/pricing-change",
"activities": [
"Produce What Customers and Sellers Read"
]
},
{
"process": "ref/mkt/product-positioning",
"activities": [
"Test the Positions against the Evidence",
"Write the Messaging"
]
},
{
"process": "ref/mkt/refresh-content",
"activities": [
"Brief the Pages in the Cycle",
"Edit the Rewrite"
]
},
{
"process": "ref/mkt/speaking-engagement",
"activities": [
"Shape the Talk",
"Rehearse the Talk",
"Reuse the Talk"
]
},
{
"process": "ref/mkt/video-asset",
"activities": [
"Settle the Script",
"Review the Cut"
]
},
{
"process": "ref/mkt/website-content-update",
"activities": [
"Edit the Change"
]
}
]
}
}
}
README.mdWhat the package is, and why it carries no mcp.json.
# editor This is an abstract agent, published as an Agent Plugin. An abstract agent is a job description rather than an implementation. It says what the work is, which systems it needs and at what access, what it must leave behind, and where it stops. Any agent that leaves the records can implement it. The plugin carries thirteen skills, no `mcp.json`, and everything the abstract agent owes under `com.agentcatalog.agent/`. ## Why there is no mcp.json A conformant server entry in `mcp.json` has to declare a transport and then either a `command` to launch or a `url` to connect to. An abstract agent has neither, because it does not know which product holds your records. Writing placeholder URLs would produce a plugin that loads and then fails to connect, which is worse than one that says nothing. Section 6.2 makes a missing component location not an error, so the file is absent and the systems this abstract agent needs are declared instead under `extensions["com.agentcatalog.agent"].servers`, as a role, an access level and what the access is for. Whoever builds an agent from this plugin writes the `mcp.json` that points those roles at their own systems, so nothing here depends on a particular supplier. ## Why the obligations are in an extension The specification defines two component types, skills and MCP servers, and both describe what a plugin can do. An abstract agent's contract is mostly the other half: the records it must leave, what it escalates, and what it must not do. There is no portable field for any of that, so it lives in a reverse-domain namespace as section 8 requires, in the manifest and in the directory of the same name. ## Where this abstract agent is used `ref/mkt/advisory-board`, `ref/mkt/announce-company-news`, `ref/mkt/award-entry`, `ref/mkt/brand-audit`, `ref/mkt/brand-guidelines`, `ref/mkt/brand-positioning`, and 18 more. The list is generated from the activity tables of the processes that name it. https://agentcatalog.com/abstract-agents/editor
skills/agree-what-the-piece-argues/SKILL.mdBefore anything is drafted, the agent settles what the piece argues, the order its sections come in, and the evidence each section rests on. Use it when running `ref/mkt/customer-case-study` and 3 other reference processes.
--- name: agree-what-the-piece-argues description: Before anything is drafted, the agent settles what the piece argues, the order its sections come in, and the evidence each section rests on. Use it when running `ref/mkt/customer-case-study` and 3 other reference processes. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Agree what the piece argues ## What it does Before anything is drafted, the agent settles what the piece argues, the order its sections come in, and the evidence each section rests on. ## Where it happens The agent does this in four activities across four reference processes. Each one names the activity as that process words it. - **Agree What the Story Argues** - `ref/mkt/customer-case-study`, activity 9 - [Produce Customer Case Study](../../../../processes/marketing/customer-case-study.md) - **Agree the Outline** - `ref/mkt/long-form-content`, activity 5 - [Produce Long-Form Content](../../../../processes/marketing/long-form-content.md) - **Settle the Script** - `ref/mkt/video-asset`, activity 2 - [Produce Video Asset](../../../../processes/marketing/video-asset.md) - **Shape the Talk** - `ref/mkt/speaking-engagement`, activity 7 - [Secure Speaking Engagement](../../../../processes/marketing/speaking-engagement.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/brief-the-piece/SKILL.md"The agent opens the work by telling everyone who will write, draw or promote the same thing at the same time: what is wanted, what is changing, and when it is due. Use it when running `ref/mkt/award-entry` and 8 other reference processes."
--- name: brief-the-piece description: "The agent opens the work by telling everyone who will write, draw or promote the same thing at the same time: what is wanted, what is changing, and when it is due. Use it when running `ref/mkt/award-entry` and 8 other reference processes." license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Brief the piece ## What it does The agent opens the work by telling everyone who will write, draw or promote the same thing at the same time: what is wanted, what is changing, and when it is due. ## Where it happens The agent does this in seven activities across nine reference processes. Each one names the activity as that process words it. - **Brief the Agents that Write to It** - `ref/mkt/brand-voice`, activity 17 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Brief the Department** - `ref/mkt/content-calendar`, activity 13 - [Plan Content Calendar](../../../../processes/marketing/content-calendar.md) - **Brief the Entry** - `ref/mkt/award-entry`, activity 5 - [Submit Award Entry](../../../../processes/marketing/award-entry.md) - **Brief the Makers** - `ref/mkt/brand-guidelines`, activity 4 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) - `ref/mkt/brand-voice`, activity 4 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - `ref/mkt/customer-newsletter`, activity 5 - [Publish Customer Newsletter](../../../../processes/marketing/customer-newsletter.md) - `ref/mkt/landing-page`, activity 4 - [Publish Landing Page](../../../../processes/marketing/landing-page.md) - **Brief the Pages in the Cycle** - `ref/mkt/refresh-content`, activity 10 - [Refresh Published Content](../../../../processes/marketing/refresh-content.md) - **Brief the Piece** - `ref/mkt/long-form-content`, activity 3 - [Produce Long-Form Content](../../../../processes/marketing/long-form-content.md) - **Brief the Story** - `ref/mkt/customer-case-study`, activity 6 - [Produce Customer Case Study](../../../../processes/marketing/customer-case-study.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/build-a-document-for-an-occasion/SKILL.mdThe agent writes and assembles the material one session or audience needs, such as a pre-read, a board pack, a decision paper with its options, a pitch note, or the notice and the answers that go out with a change. Use it when running `ref/mkt/advisory-board` and 4 other reference processes.
--- name: build-a-document-for-an-occasion description: The agent writes and assembles the material one session or audience needs, such as a pre-read, a board pack, a decision paper with its options, a pitch note, or the notice and the answers that go out with a change. Use it when running `ref/mkt/advisory-board` and 4 other reference processes. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Build a document for an occasion ## What it does The agent writes and assembles the material one session or audience needs, such as a pre-read, a board pack, a decision paper with its options, a pitch note, or the notice and the answers that go out with a change. ## Where it happens The agent does this in six activities across five reference processes. Each one names the activity as that process words it. - **Build the Materials** - `ref/mkt/advisory-board`, activity 7 - [Convene Customer Advisory Board](../../../../processes/marketing/advisory-board.md) - **Build the Pack** - `ref/mkt/brief-executive-leadership`, activity 12 - [Brief Executive Leadership](../../../../processes/marketing/brief-executive-leadership.md) - **Make What the Path Needs** - `ref/mkt/onboarding-email-program`, activity 6 - [Run Onboarding Email Program](../../../../processes/marketing/onboarding-email-program.md) - **Produce What Customers and Sellers Read** - `ref/mkt/pricing-change`, activity 6 - [Launch Pricing Change](../../../../processes/marketing/pricing-change.md) - **Produce the Pitch Materials** - `ref/mkt/pitch-earned-media`, activity 6 - [Pitch Earned Media](../../../../processes/marketing/pitch-earned-media.md) - **Write the Decision Papers** - `ref/mkt/brief-executive-leadership`, activity 11 - [Brief Executive Leadership](../../../../processes/marketing/brief-executive-leadership.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/define-the-voice/SKILL.md"The agent writes down how the organization sounds: the traits it writes with, how the tone changes by channel and by situation, and the words to use, the words to avoid, and the names and spellings it settles on. Use it when running `ref/mkt/brand-positioning` and `ref/mkt/brand-voice`."
--- name: define-the-voice description: "The agent writes down how the organization sounds: the traits it writes with, how the tone changes by channel and by situation, and the words to use, the words to avoid, and the names and spellings it settles on. Use it when running `ref/mkt/brand-positioning` and `ref/mkt/brand-voice`." license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Define the voice ## What it does The agent writes down how the organization sounds: the traits it writes with, how the tone changes by channel and by situation, and the words to use, the words to avoid, and the names and spellings it settles on. ## Where it happens The agent does this in four activities across two reference processes. Each one names the activity as that process words it. - **Draft the Voice** - `ref/mkt/brand-voice`, activity 5 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Set the Tone for Each Situation** - `ref/mkt/brand-voice`, activity 6 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Set the Voice Rules** - `ref/mkt/brand-positioning`, activity 12 - [Define Brand Positioning](../../../../processes/marketing/brand-positioning.md) - **Write the Word List** - `ref/mkt/brand-voice`, activity 10 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/edit-a-draft/SKILL.mdThe agent reads a draft against what was asked for and says item by item what has to be rewritten, in terms the writer can work through without coming back to ask what was meant. Use it when running `ref/mkt/announce-company-news` and 10 other reference processes.
--- name: edit-a-draft description: The agent reads a draft against what was asked for and says item by item what has to be rewritten, in terms the writer can work through without coming back to ask what was meant. Use it when running `ref/mkt/announce-company-news` and 10 other reference processes. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Edit a draft ## What it does The agent reads a draft against what was asked for and says item by item what has to be rewritten, in terms the writer can work through without coming back to ask what was meant. It holds the piece at that standard for as many rounds as it takes. ## Where it happens The agent does this in eight activities across eleven reference processes. Each one names the activity as that process words it. - **Check It Once** - `ref/mkt/feature-release`, activity 5 - [Announce Feature Release](../../../../processes/marketing/feature-release.md) - **Edit the Card** - `ref/mkt/competitive-battlecard`, activity 8 - [Refresh Competitive Battlecard](../../../../processes/marketing/competitive-battlecard.md) - **Edit the Change** - `ref/mkt/website-content-update`, activity 5 - [Update Website Content](../../../../processes/marketing/website-content-update.md) - **Edit the Draft** - `ref/mkt/customer-case-study`, activity 11 - [Produce Customer Case Study](../../../../processes/marketing/customer-case-study.md) - `ref/mkt/customer-newsletter`, activity 8 - [Publish Customer Newsletter](../../../../processes/marketing/customer-newsletter.md) - `ref/mkt/landing-page`, activity 6 - [Publish Landing Page](../../../../processes/marketing/landing-page.md) - `ref/mkt/long-form-content`, activity 7 - [Produce Long-Form Content](../../../../processes/marketing/long-form-content.md) - **Edit the Entry** - `ref/mkt/award-entry`, activity 9 - [Submit Award Entry](../../../../processes/marketing/award-entry.md) - **Edit the Release** - `ref/mkt/announce-company-news`, activity 7 - [Announce Company News](../../../../processes/marketing/announce-company-news.md) - **Edit the Rewrite** - `ref/mkt/refresh-content`, activity 13 - [Refresh Published Content](../../../../processes/marketing/refresh-content.md) - **Review the Cut** - `ref/mkt/video-asset`, activity 8 - [Produce Video Asset](../../../../processes/marketing/video-asset.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/plan-the-slate/SKILL.mdThe agent works out how much work the makers can carry this period and lays the slate out on a schedule, with a format, an agent and a date against each piece. Use it when running `ref/mkt/content-calendar`.
--- name: plan-the-slate description: The agent works out how much work the makers can carry this period and lays the slate out on a schedule, with a format, an agent and a date against each piece. Use it when running `ref/mkt/content-calendar`. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Plan the slate ## What it does The agent works out how much work the makers can carry this period and lays the slate out on a schedule, with a format, an agent and a date against each piece. ## Where it happens The agent does this in two activities across one reference process. Each one names the activity as that process words it. - **Draft the Calendar** - `ref/mkt/content-calendar`, activity 10 - [Plan Content Calendar](../../../../processes/marketing/content-calendar.md) - **Size the Production Capacity** - `ref/mkt/content-calendar`, activity 7 - [Plan Content Calendar](../../../../processes/marketing/content-calendar.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/read-what-is-already-published/SKILL.mdThe agent reads a body of live material and says how it reads as a whole, whether it sounds and looks like one organization, and how it would fare against rules that have not issued yet. Use it when running `ref/mkt/brand-audit` and `ref/mkt/brand-voice`.
--- name: read-what-is-already-published description: The agent reads a body of live material and says how it reads as a whole, whether it sounds and looks like one organization, and how it would fare against rules that have not issued yet. Use it when running `ref/mkt/brand-audit` and `ref/mkt/brand-voice`. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Read what is already published ## What it does The agent reads a body of live material and says how it reads as a whole, whether it sounds and looks like one organization, and how it would fare against rules that have not issued yet. ## Where it happens The agent does this in three activities across two reference processes. Each one names the activity as that process words it. - **Read How the Company Sounds Now** - `ref/mkt/brand-voice`, activity 2 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Read the Set as a Whole** - `ref/mkt/brand-audit`, activity 8 - [Conduct Brand Audit](../../../../processes/marketing/brand-audit.md) - **Score Old Material against the Draft Rules** - `ref/mkt/brand-voice`, activity 12 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/rehearse-a-live-delivery/SKILL.mdThe agent runs a talk or a briefing before the real audience hears it, asks the hard questions early, runs it to time, and says what has to change. Use it when running `ref/mkt/brief-executive-leadership` and `ref/mkt/speaking-engagement`.
--- name: rehearse-a-live-delivery description: The agent runs a talk or a briefing before the real audience hears it, asks the hard questions early, runs it to time, and says what has to change. Use it when running `ref/mkt/brief-executive-leadership` and `ref/mkt/speaking-engagement`. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Rehearse a live delivery ## What it does The agent runs a talk or a briefing before the real audience hears it, asks the hard questions early, runs it to time, and says what has to change. ## Where it happens The agent does this in two activities across two reference processes. Each one names the activity as that process words it. - **Rehearse the Briefing** - `ref/mkt/brief-executive-leadership`, activity 15 - [Brief Executive Leadership](../../../../processes/marketing/brief-executive-leadership.md) - **Rehearse the Talk** - `ref/mkt/speaking-engagement`, activity 10 - [Secure Speaking Engagement](../../../../processes/marketing/speaking-engagement.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/rewrite-the-standing-wording/SKILL.mdThe agent updates the names, boilerplate and descriptions a change has made wrong, wherever the organization uses them. Use it when running `ref/mkt/brand-refresh`.
--- name: rewrite-the-standing-wording description: The agent updates the names, boilerplate and descriptions a change has made wrong, wherever the organization uses them. Use it when running `ref/mkt/brand-refresh`. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Rewrite the standing wording ## What it does The agent updates the names, boilerplate and descriptions a change has made wrong, wherever the organization uses them. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Rewrite What the Words Say** - `ref/mkt/brand-refresh`, activity 14 - [Roll Out Brand Refresh](../../../../processes/marketing/brand-refresh.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/test-a-claim-against-its-evidence/SKILL.mdThe agent checks whether the evidence under a statement carries what the statement claims, and names the gap where it does not. Use it when running `ref/mkt/brand-positioning` and `ref/mkt/product-positioning`.
--- name: test-a-claim-against-its-evidence description: The agent checks whether the evidence under a statement carries what the statement claims, and names the gap where it does not. Use it when running `ref/mkt/brand-positioning` and `ref/mkt/product-positioning`. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Test a claim against its evidence ## What it does The agent checks whether the evidence under a statement carries what the statement claims, and names the gap where it does not. ## Where it happens The agent does this in one activity across two reference processes. Each one names the activity as that process words it. - **Test the Positions against the Evidence** - `ref/mkt/brand-positioning`, activity 7 - [Define Brand Positioning](../../../../processes/marketing/brand-positioning.md) - `ref/mkt/product-positioning`, activity 9 - [Develop Product Positioning](../../../../processes/marketing/product-positioning.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/turn-a-talk-into-other-pieces/SKILL.md"The agent takes a talk that has been given and works it into the pieces that outlive it: the clips, the written version, and the slides that get shared. Use it when running `ref/mkt/speaking-engagement`."
--- name: turn-a-talk-into-other-pieces description: "The agent takes a talk that has been given and works it into the pieces that outlive it: the clips, the written version, and the slides that get shared. Use it when running `ref/mkt/speaking-engagement`." license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Turn a talk into other pieces ## What it does The agent takes a talk that has been given and works it into the pieces that outlive it: the clips, the written version, and the slides that get shared. ## Where it happens The agent does this in one activity across one reference process. Each one names the activity as that process words it. - **Reuse the Talk** - `ref/mkt/speaking-engagement`, activity 13 - [Secure Speaking Engagement](../../../../processes/marketing/speaking-engagement.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/write-rules-that-can-be-scored/SKILL.mdThe agent turns a standard into numbered rules written tightly enough that a scoring agent can apply them, each one with a passing and a failing example. Use it when running `ref/mkt/brand-guidelines` and `ref/mkt/brand-voice`.
--- name: write-rules-that-can-be-scored description: The agent turns a standard into numbered rules written tightly enough that a scoring agent can apply them, each one with a passing and a failing example. Use it when running `ref/mkt/brand-guidelines` and `ref/mkt/brand-voice`. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Write rules that can be scored ## What it does The agent turns a standard into numbered rules written tightly enough that a scoring agent can apply them, each one with a passing and a failing example. ## Where it happens The agent does this in two activities across two reference processes. Each one names the activity as that process words it. - **Turn the Voice into Scorable Rules** - `ref/mkt/brand-voice`, activity 11 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Write the Rules** - `ref/mkt/brand-guidelines`, activity 5 - [Publish Brand Guidelines](../../../../processes/marketing/brand-guidelines.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
skills/write-the-positioning-and-the-messaging/SKILL.mdThe agent writes the statement of what the organization or the product is for, the promise behind it, the proof points under it, and the version each audience gets. Use it when running `ref/mkt/brand-positioning`, `ref/mkt/brand-voice` and `ref/mkt/product-positioning`.
--- name: write-the-positioning-and-the-messaging description: The agent writes the statement of what the organization or the product is for, the promise behind it, the proof points under it, and the version each audience gets. Use it when running `ref/mkt/brand-positioning`, `ref/mkt/brand-voice` and `ref/mkt/product-positioning`. license: CC-BY-4.0 metadata: agent: editor agent-version: "1" --- # Write the positioning and the messaging ## What it does The agent writes the statement of what the organization or the product is for, the promise behind it, the proof points under it, and the version each audience gets. ## Where it happens The agent does this in three activities across three reference processes. Each one names the activity as that process words it. - **Build the Messaging Architecture** - `ref/mkt/brand-voice`, activity 7 - [Define Brand Voice and Messaging](../../../../processes/marketing/brand-voice.md) - **Write the Messaging** - `ref/mkt/product-positioning`, activity 12 - [Develop Product Positioning](../../../../processes/marketing/product-positioning.md) - **Write the Positioning and the Promise** - `ref/mkt/brand-positioning`, activity 10 - [Define Brand Positioning](../../../../processes/marketing/brand-positioning.md) ## What to record Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason. That contract covers every activity this abstract agent takes on, and it is repeated in `com.agentcatalog.agent/RECORDS.md`. What the abstract agent does not do is in `com.agentcatalog.agent/NOT.md`.
mcp.jsonNot in the package.
Every server entry in an mcp.json needs either a command to run or a url to connect to, and the package cannot know either of those until somebody points it at their own systems. So the file is left out, the systems the agent needs are named by role in plugin.json, and you write the mcp.json when you set it up.
Setting it up
Point it at your systems
To run an agent built from this package, write an mcp.json that says which of your systems fills each role. The package names these 4 roles, and the access level beside each one is the most the work needs.
asset-store · writestyle-guide-and-the-house-standard · readbrief-and-the-approved-outline · readsource-list-the-draft-cites · read
mcp.json
You write this file. It is not part of the package.
{
"$schema": "https://agent-plugins.org/schemas/1.1.0/mcp.schema.json",
"mcpServers": {
"asset-store": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/asset-store"
},
"style-guide-and-the-house-standard": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/style-guide-and-the-house-standard"
},
"brief-and-the-approved-outline": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/brief-and-the-approved-outline"
},
"source-list-the-draft-cites": {
"type": "streamable-http",
"url": "https://mcp.yourcompany.com/source-list-the-draft-cites"
}
}
}
The hostnames here are a placeholder for your own. Whichever product holds each of these systems, the agent you build is yours and nobody else has to be involved in running it.
Where it works
24 processes, 51 activities
The highlighted activities are the ones this abstract agent does. The other activities in each process belong to other abstract agents or to people.
Convene Customer Advisory Boardref/mkt/advisory-board · 16 activities
Build the Materials
Announce Company Newsref/mkt/announce-company-news · 17 activities
Edit the Release
Submit Award Entryref/mkt/award-entry · 16 activities
Brief the Entry · Edit the Entry
Conduct Brand Auditref/mkt/brand-audit · 17 activities
Read the Set as a Whole
Publish Brand Guidelinesref/mkt/brand-guidelines · 16 activities
Brief the Makers · Write the Rules
Define Brand Positioningref/mkt/brand-positioning · 16 activities
Test the Positions against the Evidence · Write the Positioning and the Promise · Set the Voice Rules
Roll Out Brand Refreshref/mkt/brand-refresh · 26 activities
Rewrite What the Words Say
Define Brand Voice and Messagingref/mkt/brand-voice · 19 activities
Read How the Company Sounds Now · Brief the Makers · Draft the Voice · Set the Tone for Each Situation · Build the Messaging Architecture · Write the Word List · Turn the Voice into Scorable Rules · Score Old Material against the Draft Rules · Brief the Agents that Write to It
Brief Executive Leadershipref/mkt/brief-executive-leadership · 20 activities
Write the Decision Papers · Build the Pack · Rehearse the Briefing
Refresh Competitive Battlecardref/mkt/competitive-battlecard · 16 activities
Edit the Card
Plan Content Calendarref/mkt/content-calendar · 15 activities
Size the Production Capacity · Draft the Calendar · Brief the Department
Produce Customer Case Studyref/mkt/customer-case-study · 21 activities
Brief the Story · Agree What the Story Argues · Edit the Draft
Publish Customer Newsletterref/mkt/customer-newsletter · 17 activities
Brief the Makers · Edit the Draft
Announce Feature Releaseref/mkt/feature-release · 10 activities
Check It Once
Publish Landing Pageref/mkt/landing-page · 15 activities
Brief the Makers · Edit the Draft
Produce Long-Form Contentref/mkt/long-form-content · 18 activities
Brief the Piece · Agree the Outline · Edit the Draft
Run Onboarding Email Programref/mkt/onboarding-email-program · 14 activities
Make What the Path Needs
Pitch Earned Mediaref/mkt/pitch-earned-media · 15 activities
Produce the Pitch Materials
Launch Pricing Changeref/mkt/pricing-change · 16 activities
Produce What Customers and Sellers Read
Develop Product Positioningref/mkt/product-positioning · 17 activities
Test the Positions against the Evidence · Write the Messaging
Refresh Published Contentref/mkt/refresh-content · 22 activities
Brief the Pages in the Cycle · Edit the Rewrite
Secure Speaking Engagementref/mkt/speaking-engagement · 16 activities
Shape the Talk · Rehearse the Talk · Reuse the Talk
Produce Video Assetref/mkt/video-asset · 15 activities
Settle the Script · Review the Cut
Update Website Contentref/mkt/website-content-update · 12 activities
Edit the Change
The records it leaves
What it writes down
Every edit at a version, saying what has to change and why, item by item, in terms the writer can work through without coming back to ask what was meant. A line for anything it flagged for another pass, naming the agent it went to. When it holds the version, the merge, the decline or the revision request it gave in every round, each with its reason.
Where it stops
What it will not do
Does not rewrite the piece. It says what has to change and the writer makes the change, which is what keeps the writer answerable for the words. Does not go and find the evidence - that is the researcher agent, and this agent's job is to notice when the evidence is missing. Does not score the piece against the brand guidelines or the claims register, because that is the brand-check agent and it runs as a separate pass. Does not choose the title, the headings or the links for findability, which is the search-optimizer agent. Does not commission the piece or set its date, which the content-manager agent does. It is not a signer: it says the piece is ready, and named people sign.
These limits are written into the job on purpose, so that everyone adopting the process knows which decisions stay with a person.
Agents built from this plugin
Who has built one, and how far each one got
Nobody has built one yet. The catalog says so rather than leaving the section off, because an empty list is a fact about the document.
declare it
implements: editor v1
One line in your agent's manifest. The catalog checks it against the records your runs actually leave.
A guess, not a claim
The closest things in the catalog, by description
Nobody above has claimed any of these. This is the nearest match by wording between the job written above and what publishers wrote about their own listings, closest first, and wording is not evidence: some of these will be wrong, and a job with nothing suitable still gets whatever was nearest. The publishers have never heard of this abstract agent and have promised nothing about it. What a claim looks like is in the table above, and it is signed.
Might be reached for
Not workers. These are capability that whatever does the job would use, which is a different relationship and kept in its own column for that reason.
- Writing Clinic MCP server · stockandawe.github.io · run it yourself · mirrored from the MCP registry Free MCP tool: paste your writing, get a one-line diagnosis and a tightened rewrite.
- Standout MCP server · standoutmcp.io · hosted · mirrored from the MCP registry Anti-slop design taste for AI coding agents: art directions, section code, 0-100 page critic.
- map MCP server · maraventano.github.io · run it yourself · mirrored from the MCP registry A subtractive MCP for agent commerce: read · pay · publish. Teacher, not butler.